A presentation that I gave recently at BioConference Live on Genetics and Genomics focused on using re-sequencing in human genetic testing. While this talk focused on technology comparison – microarrays vs NGS, there were a few important aspects that are independent of the platform used for re-sequencing. A critically important component of re-sequencing is availability of reference materials, in particular, human genetic material and human genome.
What is the difference? A reference human genetic material is critical for development, validation and QC of genetic testing assays, while the human genome reference is critical for analysis of the data as well as assay design. There are two groups that attempt to assist the community with those needs.
First is the Genetic Testing Reference Materials Coordination Program (GeT-RM), the second is the Genome Reference Consortium. GeT-RM is aiming to assist in process of development, characterization of reference materials and subsequently in their distrtibution. They also facilitate information exchange between users and providers of reference materials. When I was working on development of genetic tests I realized very quickly that such materials are critical for the process. There is no way to properly develop an assay by using a mix of random DNA samples from the freezer archives. Fortunately explaining this issue was never a problem, most laboratories already knew how to obtain and use reference materials listed on GeT-RM site. A more complicated issue was which version of the human genome reference to use. Documenting the version used in development of assay is important, then documenting the version used in analysis is even more important.
Why is it such a big deal? Human genome has been sequenced, we just celebrated the anniversary this year. However, what we refer to as a reference genome is really, as Deanne Church said, a reference assembly. And when you think back to the days of sequencing of human genome, you quickly realize that what we now consider a reference is in fact a mix of a number of DNA samples from various people. Moreover it contains some disease causing mutations, for example in BRCA1, which if not identified correctly will not flag that you have a mutation. This reference assembly is used for primer/probe design, comparison of your new sequence to identify mutations or new variants in your sample. While we can mask known SNPs and repeats in the genome for primer/probe design we have no guarantees that we catch them all at the design stage. It is also clear now from NGS that human genome is not fully sequenced, there are regions that have very low to no coverage. So improved technology did not really translate as yet to 100% sequence. That said we are now able to fast re-sequence not just a panel of genes or exome but the entire genome in just a few days, capturing probably most of the important information.
What then should be the reference genome? An excellent discussion on the topic was recently presented by Nathan Pearson at the Clinical Genome Conference in San Francisco and subsequently in his blog. I will not go into the details of his discussion but it is critical to realize that we need a reference genome in order to do re-sequencing. The proposal is to use human ancestral genome as a reference, thus creating hopefully a much better baseline for everyone else’s genome sequence.
Ok, so my DNA was sequenced, I have a good reference genome, what next? Traditional approach was to tell use what is different from the reference. We often asked ourselves is my DNA normal? Do I have a mutation? It is currently quite common to report a genotype for a person across a particular gene or genes. Yet, variant reporting even from re-sequencing experiments is still present. Again, Nathan in his blog goes into great depth why this is a problem and I am not going to repeat his arguments here. I want to focus on a bit different angle here.
Why do we sequence the whole genome? Because we can, because we need to find the mutation responsible for a rare disease, because we want to know why we are the way we are, or because we want to know if we are at risk for a cardiovascular disease, cancer etc. Many of you used Counsyl or 23 and me to have your risks assessed. In one of my previous blogs I said that those tests are only looking at one side of the coin, at a small selection of genes. We still have no idea how all the variants present in our genome affect the final phenotype. I can hear those saying … but there is environment too, let us focus just on the genome for now. Let’s take breast cancer, presence of particular mutations in BRCA1 or BRCA2 is poor prognosis, yet it does not mean that in 100% of cases breast cancer will develop and when our gene does not have any mutations we are not immune from developing breast cancer. If our entire genome has been sequenced we can attempt to assess other genes, whose products interact with BRCA proteins or modify their actions and see if we have any protective mutations present. This way of looking at whole genome can explain some of the contradictory findings in literature regarding a particular polymorphism causing one disease but protecting you from another.
Now let’s look at a favorite drink of scientists – coffee. Probably the most controversial drink used by millions of people worldwide. It has been claimed to be bad for those with high blood pressure, can extend your life but not if you are under 55 years old, heavy drinking is now claimed to reduce the risk of prostate cancer relapse and finally it can have mixed effects on the same condition, it was said to both cause and improve depression.
So where is the truth in all this? If you consider the fact that coffee has to be metabolized, any change in the metabolic pathway or the regulators of the pathway, including transcriptional factors, protein modifiers etc, will affect our organism’s relationship with coffee. The only way to sort this out is to look at a lot of genotypes derived from whole genome sequencing and identify what other factors might be involved in our response to coffee.
Leave a comment